Daily Analysis of COVID-19 Data from Johns Hopkins University¶
An open source resource to support interpretation of COVID-19 case data.
Josef Kellndorfer, Ph.D., Earth Big Data, LLC; Richard Signell, Ph.D.
Click here to share this website.
CAUTION regarding the interpretation of Johns Hopkins COVID-19 data:
The daily data provided by Johns Hopkins tend to contain some errors, often fixed the next day. See Data Sources section for community feedback.
More and more studies estimate COVID-19 detection reported in daily data to be a small fraction (as low as 5 to 6 percent) of confirmed cases and fatalities. Thus, critical interpretation of the daily Johns Hopkins data is advised when assessing the scale of the COVID-19 pandemic. Please see the Suggested Reading/Watching section.
NOTE: You can hover over plots to see actual numbers. For more details see Interacting with the plots.
News 2020-04-09¶
- Known issue with NY Deaths allocation
- Added Massachusetts Counties time series plots
- Added Massachusetts Counties ranking per Capita
Content¶
Introduction: Population Normalization and Logarithmic Scaling
Interacting with the plots
Examples Linear and Logarithmic Scale
Confirmed Cases: USA
Confirmed Cases: Germany
Latest Data: Confirmed Cases and Deaths by Country
Latest Top 12 Countries Total Cases and Deaths
Latest per Capita Confirmed Cases and Deaths
Timelines: Countries
Confirmed Cases: Country Comparison
Deaths: Country Comparison
Timelines: U.S. States
Confirmed Cases: U.S. States Comparison
Deaths: U.S States Comparison
Timelines: Cape Cod and Islands
Cape Cod and Islands Confirmed Cases
Cape Cod and Islands Deaths
Mortality
Latest Mortality Rate
Mortality Timeline
3-day Change by Country
3-day Change in Confirmed Cases
3-day Change in Deaths
Doubling Rates Countries
Doubling Rate in Days: Confirmed Cases
Doubling Rate in Days: Deaths
Doubling Rates U.S.
Doubling Rate in Days: Confirmed Cases (U.S.)
Doubling Rate in Days: Deaths (U.S)
Selected U.S. Counties
Selected Counties: Confirmed Cases
Selected Counties: Deaths
Timelines: Massachusetts Counties
MA Counties: Confirmed Cases per Capita
MA Counties: Deaths per Capita
Introduction: Population Normalization and Logarithmic Scaling¶
back to top
These plots show the daily status of COVID-19 cases as reported by Johns Hopkins University. We want to caveat the data analysis and interpretation:
- Keep in mind that data are changing frequently as more Covid-19 testing (for cases and deaths) becomes more readily available globally.
- Overall: Use your own judgement on the trustworthiness on reported data, i.e. where the data come from. (see section on Data Sources)
We chose to plot totals and numbers normalized by population (expressed as per 100,000). Also, it is advantageous to plot case totals (confirmed infections, deaths) in logarithmic scale where trends and parallels between countries become more obvious. Note, that a straight line trending upwards in logarithmic scale indicates exponential increase! Taking a close look at the plots, one will discern differences and similarities, and that for the most part initial stages are similar in all countries with a time lag. What to look out for is whether the measures taken by countries, foremost social distancing show the desired effects of slowing and eventually levelling out the exponential upwards trends. We produce plots for confirmed cases and deaths, which may be somewhat more reliable with respect to an impact for a country while tests are rolled out in larger numbers. We also plot mortality rates, 3-day change curves and doubling rates of confirmend cases and deaths.
This is work in progress, stay tuned.
You can get the notebook underlying this work at: https://github.com/EarthBigData/covid19
Interacting with the plots¶
Hover: See actual numbers when hovering over a plot.
Control buttons: Interact with the plots: Pan, Zoom in/out, Reset, Save.
Labels: In the legend, click on label to dim/highlight a specific country or state.
Cape Cod and Islands¶
The plots below show the confirmed cases and deaths over time on Cape Cod and the Islands compared to the trends in Massachusetts and the U.S.. To compare, the plots show numbers normalized by population (based on U.S. Census population data for 2019). The plots are shown as normalized totals and in logarithmic scale.
Cape Cod and Islands Confirmed Cases¶
Mortality Rates Country Comparison¶
The plot below shows the mortality rate in percent computed as:
$Mortality=\frac{Deaths}{Infected} * 100$
A couple of caveats:
- Because there is a timelag from infection to death, the rates computed on a daily basis might be underestimating the rate.
- Underreporting of COVID-19 deaths is recognized by many countries with inadequate tallying systems.
- The rates might be vastly overestimated because the lag in testing is widespread. South Korea has done the most intensive testing, hence South Korea data seem most reliable for mortality assessment.
3-Day Change in Confirmed Cases: Country Comparison¶
The plots below show the change of total number of confirmed cases compared to three days before the plotted date. A factor 2 means the cases doubled after three days. A factor 1 means no new confirmed cases are reported compared to three days before. (Plots also inspired by Jennifer Bardwell, Jim Bardwell).
3-Day Change in Deaths: Country Comparison¶
The plots below show the change of total number of deaths compared to three days before the plotted date. A factor 2 means the cases doubled after three days. A factor 1 means no new deaths are reported compared to three days before. (Plots also inspired by Jennifer Bardwell, Jim Bardwell).
Number of Days to Double Confirmed Cases: Country Comparison¶
The plots below show the change rate (${doubling.rate}_{confirmed.cases}$) in number of days for confirmed cases to double ($days_{confirmed.cases.double}$). This is expressed as
${doubling.rate}_{confirmed.cases} = \frac{1}{days_{confirmed.cases.double}}$
In this representation a factor of 1 means cases double every day, 0.5 means cases double every 2nd day, 0.33 means cases double every third daty, 0.25 menas cases double every 4th day, etc. When the line approaches 0, no more cases are identified.
Plots begin at more than 100 confirmed cases.
(Plots inspired by Jennifer Bardwell, Jim Bardwell).
Number of Days to Double Deaths: Country Comparison¶
The plots below show the change rate (${doubling.rate}_{deaths}$) in number of days for deaths to double ($days_{deaths.double}$). This is expressed as
${doubling.rate}_{deaths} = \frac{1}{days_{deaths.double}}$
In this representation a factor of 1 means death counts double every day, 0.5 means death counts double every 2nd day, 0.33 means death counts double every third day, 0.25 menas death counts double every 4th day, etc. When the line approaches 0, no more deaths are counted.
Plots begin at more than 25 deaths.
(Plots inspired by Jennifer Bardwell, Jim Bardwell).
Number of Days to Double Confirmed Cases: U.S. States Comparison¶
The plots below show the change rate (${doubling.rate}_{confirmed.cases}$) in number of days for confirmed cases to double ($days_{confirmed.cases.double}$). This is expressed as
${doubling.rate}_{confirmed.cases} = \frac{1}{days_{confirmed.cases.double}}$
In this representation a factor of 1 means cases double every day, 0.5 means cases double every 2nd day, 0.33 means cases double every third daty, 0.25 menas cases double every 4th day, etc. When the line approaches 0, no more cases are identified.
Plots begin at more than 100 confirmed cases.
(Plots inspired by Jennifer Bardwell, Jim Bardwell).
Number of Days to Double Deaths: U.S. States Comparison¶
The plots below show the change rate (${doubling.rate}_{deaths}$) in number of days for deaths to double ($days_{deaths.double}$). This is expressed as
${doubling.rate}_{deaths} = \frac{1}{days_{deaths.double}}$
In this representation a factor of 1 means death counts double every day, 0.5 means death counts double every 2nd day, 0.33 means death counts double every third day, 0.25 menas death counts double every 4th day, etc. When the line approaches 0, no more deaths are counted.
Plots begin at more than 10 deaths.
(Plots inspired by Jennifer Bardwell, Jim Bardwell).
Massachusetts by County¶
Source:https://en.wikipedia.org/wiki/File:Massachusetts-counties-map.gif
Confirmed Cases¶
We hope these data are informative and convey how seriously we have to take the COVID-19 pandemic. Stay safe.¶
Suggested Reading/Watching:¶
(2020-04-06) COVID-19: on average only 6% of actual SARS-CoV-2 infections detected worldwide
(2020-04-05) Coronavirus death toll: Americans are almost certainly dying of Covid-19 but being left out of the official count
(2020-03-27) Watch Interview with Prof. Kim Woo-Ju, Korea University, College of Medicine
(2020-03-19) Coronavirus Interview with Larry Brilliant
(2007-04-08) Severe Acute Respiratory Syndrome Coronavirus as an Agent of Emerging and Reemerging Infection (Concluding remarks of this paper: The presence of a large reservoir of SARS-CoV-like viruses in horseshoe bats, together with the culture of eating exotic mammals in southern China, is a time bomb. The possibility of the reemergence of SARS and other novel viruses from animals or laboratories and therefore the need for preparedness should not be ignored.) GOT SCIENCE?
Data Sources¶
Johns Hopkins University daily updated COVID-19 data¶
COVID-19 confirmed cases, deaths and recovered cases data are streamed from the The Center for Systems Science and Engineering (CSSE) at Johns Hopkins University. The CCSE COVID-19 GitHub Repo has more information about these data and their sources.
Community feedback on detected errors in the daily data can be accessed here.
UN Population Data¶
We obtain the Population data from UN statistics. UN Population Data Sets have more information about these data and their sources.
US Population Data¶
US population data ar obtained from US Census statistics. US Population Data Sets have more information about these data and their sources.